Tax-Loss Harvesting Study 2021—Which Robos Took Your Tax Bill Seriously?

Posted on May 24, 2021

  • Not all robos implemented tax-loss harvesting to the same extent. While some realized net losses of over 8% of the account value on the year, others did not realize any losses at all
  • TD Essential Portfolios and Schwab Intelligent Portfolios were top choices for tax-loss harvesting over the year
  • SigFig, UBS, and Citizens were at the bottom of the pack with virtually no net realized losses on the year

Tax-Loss Harvesting Study 2021—Which Robos Took Your Tax Bill Seriously?

The large sell-off at the beginning of the global pandemic has provided the opportunity to assess the efficacy of tax-loss harvesting services at major robo advisors. Backend Benchmarking tracks a set of specific accounts that offer tax-loss harvesting which were all opened and funded at the same time and receive monthly deposits (note that these accounts are different from the main accounts that we use to measure performance in the Robo Report). As of the end of 2020, the insight is clear: not all tax-loss harvesting services are made equally. TD Essential Portfolios and Schwab Intelligent Portfolios stand out for the highest percentage of realized net losses for the year. As of the end of 2020, TD realized over 9.5% of its portfolio in net short-term losses, while Schwab realized over 8.3%. Meanwhile, our tax-loss harvesting study group of 11 robo advisors averaged 3.59% realized net losses as a percentage of year-end value.

TD and Schwab also demonstrated their aggressiveness of harvesting losses by heavy portfolio turnover. TD experienced a 120% portfolio turnover, while Schwab showed 116%. On the other hand, the average of the portfolios analyzed in our tax-loss harvesting study group experienced turnover just under 50%. The benefits of a more aggressive tax-loss harvesting strategy accrue to the investor’s bottom line. Short-term losses not only offset gains but can also be used to offset some ordinary income in many cases, both of which bolster the after-tax performance of these leading robos.

Which robos did not take TLH seriously in the age of the Pandemic?

Towards the middle of the pack, Wealthfront realized net losses of just under 5% of its total portfolio, while Wells Fargo realized 3.75%. Both of these portfolios were fairly active, with Wealthfront standing out with a portfolio turnover rate of 87%. Morgan Stanley Access Investing showed some signs of tax-loss harvesting with 2.14% of its portfolio in net realized losses. At the bottom of the pack were SigFig, Citizens Bank SpeciFi, and UBS. SigFig did not realize any losses, while SpeciFi, a robo advisor backed by SigFig’s technology, only realized 0.05% net losses in the year. This is in light of estimated unrealized net losses above 9% of the account value for both accounts at the end of the first quarter of 2020. Similarly, UBS Advice Advantage, another SigFig-backed robo advice product, reported only 0.13% net losses on the year. Given that 2020 included a quarter when stock indices were down over 30% from market highs, investors would expect more opportunities to sell struggling asset classes to harvest losses for later use. But this was not the case for these three accounts.

A reminder on tax-loss harvesting:

Ultimately, when selecting a robo advisor for its ability to execute tax-loss harvesting, it is important to be aware that implementations vary widely. For some, it seems to be a major feature, while others may be struggling to find the right algorithm to harvest losses. It is also important for investors to keep in mind that while tax-loss harvesting can improve after-tax returns, the benefits can often appear greater than they are. When a tax loss is harvested, the resulting portfolio has a lower cost basis, so realized gains will be higher when assets are sold in the future. Additionally, as markets rise, the opportunities to harvest losses fall. Tax-loss harvesting opportunities occur more frequently in newer accounts, since older accounts are more likely to hold assets which have already significantly appreciated above their original cost basis.

Posted in Performance
Tagged , , , , , , ,